{"id":"W3200657680","doi":"10.26434/chemrxiv.9758600.v1","title":"Comparison Between SMILES-Based Differential Neural Computer and Recurrent Neural Network Architectures for De Novo Molecule Design","year":2019,"lang":"en","type":"preprint","venue":"ChemRxiv","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"AstraZeneca (Canada)","funders":"European Commission","keywords":"Recurrent neural network; Computer science; Differentiable function; Artificial intelligence; Artificial neural network; Deep learning; Differential (mechanical device); Deep neural networks; Machine learning","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009950768,0.0005581657,0.0005779584,0.0005311931,0.0001840899,0.00056385,0.001072544,0.0007314277,0.002456449],"category_scores_gemma":[0.001825121,0.0002076196,0.000455368,0.00050905,0.0002576711,0.001060994,0.0005108896,0.0008528202,0.0005991959],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000944372,"about_ca_system_score_gemma":0.0009325347,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004049779,"about_ca_topic_score_gemma":0.006408614,"domain_scores_codex":[0.9996386,0.0001036008,0.0000240308,0.00005974381,0.0001326764,0.00004143356],"domain_scores_gemma":[0.9993325,0.0002517605,0.00004314427,0.0001195137,0.00020735,0.00004571151],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003832389,0.0001272435,0.0008898268,0.0002131529,0.00009235635,0.00005977772,0.00002749517,0.8705502,0.003304174,0.006689963,0.00196242,0.1157002],"study_design_scores_gemma":[0.0000121867,0.0001120397,0.0001751963,0.000007681983,0.000009568629,0.00001200097,0.000004899918,0.9964203,0.001540911,0.0009022963,0.000798584,0.00000443071],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5041504,0.01328444,0.4354976,0.00136405,0.0004080151,0.0001846629,0.0009768461,0.005597953,0.03853609],"genre_scores_gemma":[0.8877286,0.002734589,0.1024389,0.0002581515,0.00004376306,0.0001122571,0.001382109,0.0002438685,0.00505782],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.004049779,"threshold_uncertainty_score":0.008217692,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0741016661691857,"score_gpt":0.3488128468433931,"score_spread":0.2747111806742074,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}